| Along with the computer science and colour input/output technology development, people’s request on the color reappearance quality is becoming higher and higher. It was found that color usually does not match well while image reappearance among different devices, and integrating color appearance model (CAM) with color management can solve the problem well. For over20years, color appearance models have evolved to the point of international standardization. However, for complex image, the traditional CAM only takes each pixel as an independent stimulus, without considering the color constancy mechanism of human eye and the effects of each pixel. As a result, the quality image representation cannot satisfy people’s requirement.As the image reappearance, we need a method to calculate the difference between source images and representation. The method can estimate the reproduction effect. The current chromatism formulas are all based on CIE chroma value. They are appropriate for evaluating the simple color piece in the same grey background. For complex image, traditional chromatism formulas can not represent the perception chromatism between images accurately because these formulas calculate chromatism average in pixels and take the average as the whole image chromatism.This paper analyzes the limitations of traditional CAM. To improve traditional CAM, this paper raises the iCAM (image color appearance model). This paper analyzes the various image color appearance phenomenon, provides detailed design and realization process of iCAM (including positive process and the reverse process) and then compares the reapereances affect with traditional CAM.In addition, this paper inproves the existing chromatism formula based on ICAM, and puts forward a new image quality assessment method based on IPT color space. Experiments show that this method is better than the traditional chromatism formula with better evaluation effect, more close to the true eye perception. |
PDF Full Text Request